You copied the Doc URL to your clipboard.

SQRDMLAH (vector)

Signed Saturating Rounding Doubling Multiply Accumulate returning High Half (vector). This instruction multiplies the vector elements of the first source SIMD&FP register with the corresponding vector elements of the second source SIMD&FP register without saturating the multiply results, doubles the results, and accumulates the most significant half of the final results with the vector elements of the destination SIMD&FP register. The results are rounded.

If any of the results overflow, they are saturated. The cumulative saturation bit, FPSR.QC, is set if saturation occurs.

Depending on the settings in the CPACR_EL1, CPTR_EL2, and CPTR_EL3 registers, and the current Security state and Exception level, an attempt to execute the instruction might be trapped.

It has encodings from 2 classes: Scalar and Vector

Scalar
(Armv8.1)

313029282726252423222120191817161514131211109876543210
01111110size0Rm100001RnRd
S

Scalar

SQRDMLAH <V><d>, <V><n>, <V><m>

if !HaveQRDMLAHExt() then UNDEFINED;

integer d = UInt(Rd);
integer n = UInt(Rn);
integer m = UInt(Rm);
if size == '11' || size == '00' then UNDEFINED;
integer esize = 8 << UInt(size);
integer datasize = esize;
integer elements = 1;
boolean rounding = TRUE;
boolean sub_op = (S == '1');

Vector
(Armv8.1)

313029282726252423222120191817161514131211109876543210
0Q101110size0Rm100001RnRd
S

Vector

SQRDMLAH <Vd>.<T>, <Vn>.<T>, <Vm>.<T>

if !HaveQRDMLAHExt() then UNDEFINED;

integer d = UInt(Rd);
integer n = UInt(Rn);
integer m = UInt(Rm);
if size == '11' || size == '00' then UNDEFINED;
integer esize = 8 << UInt(size);
integer datasize = if Q == '1' then 128 else 64;
integer elements = datasize DIV esize;
boolean rounding = TRUE;
boolean sub_op = (S == '1');

Assembler Symbols

<V> Is a width specifier, encoded in size:
size <V>
00 RESERVED
01 H
10 S
11 RESERVED
<d>

Is the number of the SIMD&FP destination register, in the "Rd" field.

<n>

Is the number of the first SIMD&FP source register, encoded in the "Rn" field.

<m>

Is the number of the second SIMD&FP source register, encoded in the "Rm" field.

<Vd>

Is the name of the SIMD&FP destination register, encoded in the "Rd" field.

<T> Is an arrangement specifier, encoded in size:Q:
size Q <T>
00 x RESERVED
01 0 4H
01 1 8H
10 0 2S
10 1 4S
11 x RESERVED
<Vn>

Is the name of the first SIMD&FP source register, encoded in the "Rn" field.

<Vm>

Is the name of the second SIMD&FP source register, encoded in the "Rm" field.

Operation

CheckFPAdvSIMDEnabled64();
bits(datasize) operand1 = V[n];
bits(datasize) operand2 = V[m];
bits(datasize) operand3 = V[d];
bits(datasize) result;
integer rounding_const = if rounding then 1 << (esize - 1) else 0;
integer element1;
integer element2;
integer element3;
integer product;
boolean sat;

for e = 0 to elements-1
    element1 = SInt(Elem[operand1, e, esize]);
    element2 = SInt(Elem[operand2, e, esize]);
    element3 = SInt(Elem[operand3, e, esize]);
    if sub_op then
        accum = ((element3 << esize) - 2 * (element1 * element2) + rounding_const);
    else
        accum = ((element3 << esize) + 2 * (element1 * element2) + rounding_const);
    (Elem[result, e, esize], sat) = SignedSatQ(accum >> esize, esize);
    if sat then FPSR.QC = '1';

V[d] = result;
Was this page helpful? Yes No